HEALTH POLICY CENTER

State Variation in Medicaid

Prescriptions for Opioid Use Disorder
from 2011 to 2018
Lisa Clemans-Cope, Victoria Lynch, Emma Winiski, and Marni Epstein
August 2019

Expanding access to effective treatment for opioid use disorder (OUD) is essential to
staunching the opioid epidemic. Access to these treatments is particularly important in
Medicaid, which covers a disproportionately large share of people with OUD (MACPAC
2017). Despite strong evidence supporting the effectiveness of pharmacotherapy for
OUD (sometimes called medication-assisted treatment), research indicates that most
people with OUD do not receive treatment (Clemans-Cope, Lynch, et al. 2019; Saloner
and Karthikeyan 2015). Three medications—buprenorphine, methadone, and
naltrexone—are approved by the Food and Drug Administration (FDA) for OUD
treatment and have well-documented benefits.1 This brief uses Medicaid State Drug
Utilization Data (SDUD) to assess buprenorphine maintenance treatment in Medicaid.
We estimated trends from 2011 to 2018 and patterns across states and the District of
Columbia, which we consider a state for this analysis.

We find the following:

◼ The number of Medicaid-covered prescriptions for buprenorphine maintenance treatment

filled or dispensed in pharmacies and providers’ offices in the United States increased almost
fivefold between 2011 and 2018, from 1.3 million to 6.2 million. In 2018, 4.8 million, or 77.2
percent, of these prescriptions were in states that expanded Medicaid under the Affordable
Care Act (ACA) by early 2014.
 ◼ The national estimate of Medicaid prescriptions for buprenorphine maintenance treatment per
1,000 enrollees ages 12 and older (the “prescribing rate”) increased from 36 in 2011 to 124 in
2018. On average, the buprenorphine prescribing rate was far lower in nonexpansion states
than in other states. Between 2011 and 2018, Medicaid prescriptions for buprenorphine
maintenance treatment per 1,000 enrollees increased from 40 to 138 among states that
expanded Medicaid by early 2014 and from 16 to 41 in nonexpansion states.
◼ We found substantial variation in states’ prescribing rates in 2018, with the highest rate, 1,210
in Vermont, being 200 times greater than the lowest rate of 5 in Arkansas. We also compare
deciles of states’ prescribing rates: Ohio, at the 90th percentile (with the fifth-highest rate), has
a rate of 438, 31 times greater than the rate of 14 in Kansas, which is at the 10th percentile
(with the fifth-lowest rate).
◼ The states with buprenorphine prescribing rates at or above the 90th percentile were Ohio
(438), Montana (588), Kentucky (662), West Virginia (827), and Vermont (1,210); those with
rates at or below the 10th percentile were Arkansas (5), Texas (8), South Dakota (11), Florida
(12), and Kansas (14).
◼ At 1,210 prescriptions for buprenorphine maintenance treatment per 1,000 Medicaid enrollees
in 2018, Vermont’s prescribing rate is 46 percent higher than the next highest rate. Though
Vermont’s higher rate likely relates to greater treatment needs than those of the nation overall,
they also likely reflect increased OUD treatment capacity and coverage under the Medicaid
expansion. Reportedly, 73 percent of Vermonters with OUD were in treatment in 2014, and by
2017, Vermont eliminated treatment wait lists in every county. Vermont’s higher prescribing
rate may also reflect higher dosing (e.g., 16 mg buprenorphine taken as two 8 mg tablets).
Because growing evidence suggests higher doses of buprenorphine (e.g., 16–32 mg) are more
efficacious than lower doses, Vermont’s higher prescribing rate could reflect clinically effective
dosing.
◼ Though state-level differences in prescribing rates may partially reflect differences in
underlying rates of OUD and prescription-writing practices, it is unlikely that such factors
account for the 31-fold state variation in prescriptions per enrollee we find in 2018. The low
buprenorphine prescribing rates in many Medicaid programs suggest states may need to take
up policy solutions addressing barriers to treatment and retention, including expanding
coverage under the ACA, increasing treatment capacity, and improving treatment
effectiveness.

Introduction
As noted, expanding access to effective OUD treatment is essential to staunching the opioid epidemic,
especially in Medicaid, which covers a disproportionately large share of people with OUD (MACPAC
2018). Despite the strong evidence supporting the effectiveness of pharmacotherapy for OUD
(sometimes called medication-assisted treatment), only about one-fifth of people with OUD receive

2 STATE VARIATION IN MEDICAI D PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18

treatment (Saloner and Karthikeyan 2015). Three medications used in OUD pharmacotherapy—
buprenorphine, methadone, and naltrexone—are FDA approved for treating OUD and have well-
documented benefits.2 This brief focuses on buprenorphine maintenance treatment.

Before the ACA, many low-income adults lacked access to affordable health insurance that covered
OUD treatment, and OUD treatment rates did not increase in the years before the ACA, even as OUD
rates rose (Saloner and Karthikeyan 2015). With the ACA, low-income adults with OUD in states that
expanded Medicaid gained coverage including at least one type of buprenorphine maintenance
treatment (Grogan et al. 2016), and coverage of more types has been growing (Miller 2018; SAMHSA
2018). In contrast, low-income uninsured adults in states that have not expanded Medicaid under the
ACA are less likely to have access to affordable OUD treatment. Research finds that Medicaid-covered
OUD treatment increased more in 2014 in states that expanded Medicaid under the ACA (Wen et al.
2017). Subsequent research shows a larger increase in Medicaid-reimbursed prescriptions and
Medicaid coverage for OUD treatment in expansion states than in other states from 2010 to 2015
(Maclean and Saloner 2017). In prior analysis of data from 2011 to 2016, we found buprenorphine
receipt and spending increased under Medicaid after 2014, particularly in states that expanded
Medicaid by early 2014 (Clemans-Cope, Lynch, et al. 2017), and these trends continued into 2017
(Clemans-Cope, Epstein, Lynch, et al. 2019).

This analysis focuses on Medicaid coverage of buprenorphine maintenance treatment, including the
combination buprenorphine-naloxone medication, where naloxone is added to buprenorphine to deter
misuse. As of October 2018, Medicaid programs covered buprenorphine for OUD in all states, and
several states expanded access to buprenorphine through policies such as eliminating prior
authorization requirements (Miller 2018). However, some state Medicaid programs have prior
authorization requirements for buprenorphine treatment or require reauthorization or documentation
of participation in counseling, and most apply dosage limits, such as a maximum daily dose of 16 or 32
mg.

In this brief, we use Medicaid SDUD to estimate trends and patterns across states in prescriptions
for buprenorphine maintenance treatment per 1,000 Medicaid enrollees from 2011 to 2018. We
examined trends overall, by state, and by states’ Medicaid expansion status and timing. This analysis
draws from methods used in a recent study of state-level variation in opioid prescribing in Medicaid
programs (Schieber et al. 2019).

Methods
We used Medicaid SDUD files from 2011 to 2018 to assess prescriptions for buprenorphine
medications FDA approved for OUD treatment. We identified OUD- and opioid overdose–related
prescriptions by linking the National Drug Code numbers in Medicaid SDUD with drug information
published by the FDA. Medicaid SDUD exclude prescriptions written by prescribers at some safety net
providers participating in the 340B medication rebate program, such as federally funded clinics
(MACPAC 2018; Murrin 2016). In addition, methadone OUD treatment, dispensed only at opioid

STATE VARIATION IN M EDICAID PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18 3

treatment programs, is not included in Medicaid SDUD. Thus, the SDUD exclude all Medicaid
reimbursements of OUD medication treatments. Because Medicaid SDUD suppress data for National
Drug Code state quarters with fewer than 11 prescriptions, we imputed the missing prescription count
data based on an analysis of one year (2014) of claims-level data from the Medicaid Statistical
Information System (MSIS).

We computed estimated 2018 buprenorphine maintenance treatment prescription counts by

aggregating data for the first three quarters of 2018 (for which revised data are available) plus an
imputed count for the fourth quarter of 2018, estimated by multiplying the third quarter of 2018
prescription count by the average percentage change over the past four quarters. We computed trends
spanning 2011 through estimated 2018 as the mean annual percentage change. We computed trends of
shorter time segments as the annual percentage change.

Following research assessing variation in states’ prescribing patterns (Schieber et al. 2019), we
quantified state-level geographic differences in buprenorphine maintenance treatment prescriptions by
comparing the lowest and highest rates estimated across states, as well as by comparing the rates for
the states at the 10th and 90th percentiles. We used the ratio between the estimate for the state at the
90th percentile and the state at the 10th percentile to assess the relative difference between states.

To confirm our findings in the Medicaid SDUD, we checked them against the most recent quarter of
MSIS enrollee-level prescription data covered by our current data usage agreement with the Centers
for Medicare & Medicaid Services.3 We examined buprenorphine maintenance treatment prescription
counts for enrollees with any buprenorphine prescription for OUD in the MSIS for six states, with a
particular focus on Vermont, which stood out in the SDUD. To inform interpretation of the SDUD
estimates, we also examined prescriptions per month using the MSIS.

Key Measures
Buprenorphine maintenance treatment prescription counts show the number of buprenorphine
prescriptions FDA approved for OUD treatment filled and dispensed in outpatient settings, such as
pharmacies, and processed as Medicaid outpatient drug claims.

The number of Medicaid enrollees ages 12 and older by state is the estimated annual Medicaid
enrollment of full-benefit enrollees ages 12 and older from 2011 to 2018. We used multiple data
sources, including person-level analytical data files of administrative records from the MSIS and the
Medicaid Analytic eXtract, to directly derive enrollment counts, using data from those in the universe of
interest.4 We adjusted counts based on relevant MSIS/Medicaid Analytic eXtract data and/or reports
states submitted to the Centers for Medicare & Medicaid Services with broader Medicaid population
counts.

Buprenorphine maintenance treatment prescriptions per 1,000 Medicaid enrollees ages 12 and older,
or the prescribing rate, divide buprenorphine maintenance treatment prescriptions by the estimated
number of Medicaid enrollees ages 12 and older (in thousands) in each state and year. We derive this

4 STATE VARIATION IN MEDICAI D PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18

rate from three factors: prevalence of OUD, treatment and retention rates, and treatment dose. In this
dataset, the share of Medicaid enrollees with OUD and the treatment rate are unknown. In addition,
treatment dose may be reflected as higher prescribing rates in this data. For example, a typical dose of
16 mg buprenorphine is commonly taken as two 8 mg tablets and could appear in the dataset as two
separate prescriptions (ASAM 2015).

The relative variation ratio measures relative state variation as the ratio of the 90th-percentile state to
the 10th-percentile state. The 90th-percentile state has the fifth-highest prescribing rate and the 10th-
percentile state has the fifth-lowest prescribing rate.

Medicaid Expansion Status and Timing

As shown in tables 1 and 2, we categorize states into four groups by the timing and status of their action
on Medicaid expansion, either through the ACA or by waiver (Sommers et al. 2013):5

◼ “Early 2014 expansion states” are the 25 states and the District of Columbia that expanded
Medicaid through the ACA or a waiver on or before April 2014; some states expanded Medicaid
under the ACA in early 2014 or had expanded eligibility for adults before the ACA.
◼ “Later 2014–16 expansion states” are the six states that expanded Medicaid between April
2014 and August 2016.
◼ “2019 expansion states” are the two states that expanded Medicaid after December 2018.
◼ “Nonexpansion states” are the 17 states that had not implemented a Medicaid expansion by
January 2019.

STATE VARIATION IN M EDICAID PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18 5

TABLE 1
States Classified by Medicaid Expansion Status

State Expansion date State Expansion date

Early 2014 expansion states Late 2014–16 expansion states
Arizonaa 1/1/2014 Alaska 9/1/2015
Arkansasa 1/1/2014 Indianaa 2/1/2015
California 7/1/2010 Louisiana 7/1/2016
Colorado 4/1/2012 Montanaa 1/1/2016
Connecticut 4/1/2010 New Hampshirea 8/15/2014
Delaware 1/1/2014 Pennsylvania 1/1/2015
District of Columbia 7/1/2010 2019 expansion states
Hawaii 1/1/2014 Maine 1/10/2019
Illinois 1/1/2014 Virginia 1/1/2019
Iowaa 1/1/2014 Nonexpansion states
Kentuckya 1/1/2014 Alabama
Maryland 1/1/2014 Florida
Massachusetts 1/1/2014 Georgia
Michigana 4/1/2014 Idahob
Minnesota 3/1/2010 Kansas
Nevada 1/1/2014 Mississippi
New Jersey 4/14/2011 Missouri
New Mexico 1/1/2014 Nebraskab
New York 1/1/2014 North Carolina
North Dakota 1/1/2014 Oklahoma
Ohio 1/1/2014 South Carolina
Oregon 1/1/2014 South Dakota
Rhode Island 1/1/2014 Tennessee
Vermont 1/1/2014 Texas
Washington 1/3/2011 Utahb
West Virginia 1/1/2014 Wisconsinc
Wyoming
Notes: a These states used Section 1115 waivers for Medicaid expansion.
b
These states approved Medicaid expansion via ballot initiatives but have yet to implement it.
c
Wisconsin did not expand Medicaid under the Affordable Care Act but has Medicaid eligibility for adults with incomes up to the
federal poverty level.

TABLE 2
Medicaid Enrollees Ages 12 and Older, 2011 to Projected 2018, by Medicaid Expansion Status
In millions

Early 2014 Late 2014–16 2019

Year Total expansion expansion expansion Nonexpansion
2011 36.2 21.4 3.1 0.8 10.8
2012 37.8 22.5 3.2 0.8 11.2
2013 39.2 23.7 3.3 0.9 11.4
2014 48.4 32.8 3.2 0.7 11.7
2015 51.4 34.9 3.8 0.8 12.0
2016 52.2 35.2 4.1 0.8 12.1
2017 52.6 35.1 4.4 0.8 12.3
Projected 2018 52.8 34.7 4.6 0.8 12.6
Source: Estimates are based on analyses of several administrative datasets, including the Medicaid Statistical Information System
and the Medicaid Analytic eXtract.

6 STATE VARIATION IN MEDICAI D PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18

Note: See the Medicaid enrollment methodology appendix for more detail (Lynch, Winiski, and Clemans-Cope 2019).

Findings
Trends in Medicaid-Funded Buprenorphine Maintenance Treatment
Prescriptions, 2011 to 2018
In 2011, 1.3 million buprenorphine maintenance treatment prescriptions were filled or dispensed in
pharmacies and providers’ offices in the United States, and by 2018, this number rose to 6.2 million
(figure 1). Most prescriptions were from states that expanded Medicaid by early 2014, accounting for
77.2 percent of the total, or an estimated 4.8 million prescriptions in 2018.

FIGURE 1
Annual Buprenorphine Maintenance Treatment Prescriptions, by State Medicaid Expansion Status,
2011 to 2018 (Estimated)
Millions

US total Early 2014 expansion Late 2014–16 expansion

2019 expansion Nonexpansion
6.2

5.0 4.8

3.9
3.7

2.8 2.9

2.2
1.8
1.5
1.3 2.0
1.2 1.5
1.0 0.7
0.9 1.0
0.4 0.5
0.2 0.2 0.2 0.3 0.5
0.2
2011 2012 2013 2014 2015 2016 2017 Estimated
2018

URBAN INSTITUTE

Source: Urban Institute analysis of Medicaid State Drug Utilization Data from the Centers for Medicare & Medicaid Services.
Notes: The Medicaid State Drug Utilization Data exclude prescriptions written by prescribers at some safety net providers
participating in the 340B medication rebate program, such as federally funded clinics. See the methodological appendices for
more details (Clemans-Cope, Epstein, Winiski, et al. 2019; Lynch, Winiski, and Clemans-Cope 2019).

Buprenorphine maintenance treatment prescriptions per 1,000 Medicaid enrollees ages 12 and
older increased across all states, from 36 in 2011 to 124 in 2018, but per enrollee prescriptions were far
lower for nonexpansion states than other groups (figure 2). In early 2014 Medicaid expansion states,

STATE VARIATION IN M EDICAID PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18 7

rates increased from 40 to 138 prescriptions per 1,000 Medicaid enrollees between 2011 and 2018
while this rate increased from 16 to 41 in nonexpansion states.

FIGURE 2
Annual Buprenorphine Maintenance Treatment Prescriptions per 1,000 Medicaid Enrollees Ages 12
and Older, by State Medicaid Expansion Status, 2011 to 2018 (Estimated)
US total Early 2014 expansion Late 2014–16 expansion
2019 expansion Nonexpansion
266

220

194
185
164
140 169
136 138
123 118
111
97 105
124
80 83
68 95
61 60
57
47
50 74
40 45
45 55
41 45 41
36
29 33
16 21 25 28
19
2011 2012 2013 2014 2015 2016 2017 Estimated
2018

URBAN INSTITUTE

Source: Urban Institute analysis of Medicaid State Drug Utilization Data from the Centers for Medicare & Medicaid Services.
Notes: The Medicaid State Drug Utilization Data exclude prescriptions written by prescribers at some safety net providers
participating in the 340B medication rebate program, such as federally funded clinics. See the methodological appendices for
more details (Clemans-Cope, Epstein, Winiski, et al. 2019; Lynch, Winiski, and Clemans-Cope 2019).

Buprenorphine maintenance treatment prescribing rates varied widely across states and
geographic areas (figure 3). In 2011, buprenorphine maintenance treatment prescriptions per enrollee
were highest in New England and parts of Appalachia, and by 2018, several additional regions across the
United States had increased prescribing rates, including the Middle Atlantic, East North Central, and
Northern Pacific states.

8 STATE VARIATION IN MEDICAI D PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18

FIGURE 3
Number of Buprenorphine Maintenance Treatment Prescriptions per 1,000 Medicaid Enrollees,
All States and the District of Columbia, 2011, 2014, and 2018

URBAN INSTITUTE
Source: Urban Institute analysis of Medicaid State Drug Utilization Data from the Centers for Medicare & Medicaid Services.

STATE VARIATION IN M EDICAID PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18 9

Notes: Prescriptions per 1,000 Medicaid enrollees ages 12 and older were derived from all buprenorphine maintenance
treatment prescriptions in Medicaid State Drug Utilization Data for each state and the District of Columbia in that year. The 2014
quartiles were used as the break points for all three maps. The Medicaid State Drug Utilization Data exclude prescriptions written
by prescribers at some safety net providers participating in the 340B medication rebate program, such as federally funded clinics.
See the methodological appendices for more details (Clemans-Cope, Epstein, Winiski, et al. 2019; Lynch, Winiski, and Clemans-
Cope 2019).

Differences across States in Buprenorphine Maintenance Treatment Prescriptions

per Enrollee, 2011–18
As shown in table 3, we found substantial variation across states’ estimated number of prescriptions for
buprenorphine maintenance treatment per 1,000 Medicaid enrollees, with the highest rate, 1,210 in
Vermont, 200 times greater than the lowest rate of 5 in Arkansas.

TABLE 3
Buprenorphine Maintenance Treatment Prescriptions per 1,000 Medicaid Enrollees Ages 12 and
Older, All States and the District of Columbia, 2011, 2014, and 2018
By 2018 deciles of prescriptions

Total BMT Prescriptions per 1,000 Medicaid Enrollees

State 2011 2014 2018 Rank
VT 590.9 784.1 1,209.8 1
WV 143.9 210.1 826.6 2
KY 92.9 28.2 662.2 3
MT 19.8 21.5 587.5 4
OH 54.9 116.4 437.7 5
ME 206.8 281.0 403.5 6
MA 175.4 172.8 384.3 7
DE 93.4 127.8 291.2 8
MD 126.2 153.6 283.5 9
IN 31.8 53.5 256.2 10
NH 81.5 147.8 246.1 11
OR 15.8 58.2 236.1 12
PA 89.0 108.7 224.3 13
VA 60.8 79.9 217.5 14
RI 93.9 155.9 207.5 15
CT 104.7 131.2 186.9 16
AK 45.1 81.8 182.8 17
WA 11.4 18.6 145.5 18
NC 33.4 70.8 139.2 19
NM 28.5 78.0 139.2 20
WI 34.4 48.8 108.5 21
LA 21.8 29.1 99.3 22
CO 7.2 21.9 97.6 23
NJ 40.5 47.9 88.2 24
MI 19.0 17.3 84.9 25
NY 45.4 49.0 82.5 26
ND 13.8 12.2 81.6 27
AL 12.4 23.1 65.1 28
AZ 22.9 12.3 64.0 29
MO 16.3 20.3 63.0 30
UT 32.3 24.7 62.3 31

10 STATE VARIATION IN MEDICAI D PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18

 Total BMT Prescriptions per 1,000 Medicaid Enrollees
State 2011 2014 2018 Rank
DC 21.7 36.3 58.4 32
SC 14.5 19.9 51.9 33
ID 13.9 18.7 49.3 34
MN 16.5 26.3 48.8 35
IL 12.3 5.6 47.3 36
TN 61.8 89.4 35.1 37
OK 9.4 16.7 33.7 38
MS 10.4 16.5 33.5 39
IA 4.5 5.8 31.4 40
HI 15.1 23.0 31.4 41
NV 4.6 8.1 29.0 42
WY 14.4 15.6 24.4 43
CA 2.1 3.3 18.5 44
NE 0.5 1.4 15.2 45
GA 3.7 10.3 14.8 46
KS 2.6 13.0 14.1 47
FL 3.8 7.2 12.3 48
SD 2.9 4.7 10.8 49
TX 4.8 6.2 8.4 50
AR 2.1 2.6 5.4 51
US 36.2 45.2 121.8
Source: Urban Institute analysis of Medicaid State Drug Utilization Data from the Centers for Medicare & Medicaid Services.
Notes: BMT = buprenorphine maintenance treatment. Horizontal lines delineate state decile groups. Buprenorphine maintenance
treatment prescriptions per 1,000 Medicaid enrollees ages 12 and older were derived from all buprenorphine maintenance
treatment prescriptions in Medicaid State Drug Utilization Data for each state and the District of Columbia in that year. The
Medicaid State Drug Utilization Data exclude prescriptions written by prescribers at some safety net providers participating in
the 340B medication rebate program, such as federally funded clinics. See the methodological appendices for more details
(Clemans-Cope, Epstein, Winiski, et al. 2019; Lynch, Winiski, and Clemans-Cope 2019).

In 2018, states in the highest decile of buprenorphine maintenance treatment prescriptions per
Medicaid enrollee were Ohio (438), Montana (588), Kentucky (662), West Virginia (827), and Vermont
(1,210); states in the lowest decile of buprenorphine maintenance treatment prescriptions per Medicaid
enrollee were Arkansas (5), Texas (8), South Dakota (11), Florida (12), and Kansas (14).

At 1,210 in 2018, the Vermont per enrollee buprenorphine maintenance treatment prescribing rate
is higher than that of all other states—and 46 percent higher than the next highest rate, in West
Virginia.6 Vermont’s early 2014 Medicaid expansion and “hub-and-spoke” treatment system expanded
treatment dramatically throughout the state (Brooklyn and Sigmon 2017). Though more research is
needed to determine whether increased treatment decreased opioid-related deaths, Vermont had the
lowest opioid-related death rate in 2017 across New England (box 1).7

We checked Vermont’s high prescribing rate in the Medicaid SDUD against enrollee-level
prescription data in 2015 MSIS claims and found that it was 57 percent higher than that of West
Virginia. The MSIS data confirmed that, on average, Vermont enrollees receiving buprenorphine
treatment prescriptions received multiple prescriptions per month. Using the same MSIS data from

STATE VARIATION IN M EDICAID PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18 11

2015, we estimated the number of buprenorphine treatment prescriptions for enrollees with any such
prescriptions was 50.2 per year for Vermont, compared with 31.8 in West Virginia.

BOX 1
Vermont’s High Rate of Buprenorphine Maintenance Treatment Prescriptions per Medicaid Enrollee
Vermont’s higher prescribing rate likely reflects several factors, the first of which is higher treatment rates.
Vermont used its early 2014 Medicaid expansion to support a new hub-and-spoke treatment system that expanded
treatment dramatically throughout the state by leveraging the expertise of methadone clinic staff (the hubs) to
expand access to buprenorphine maintenance treatment integrated into the primary care health care system
(spokes; Clemans-Cope, Wishner, et al. 2017).a As more people sought treatment in Vermont, the state expanded
treatment so that waiting periods for treatment did not increase, and consequently, Vermont now has the highest
capacity for treating OUD in the United States (Brooklyn and Sigmon 2017). With a treatment rate of 73 percent in
2014, when a major treatment expansion began (Brooklyn and Sigmon 2017), current treatment rates could exceed
80 percent. In late 2017, Governor Scott announced that the state had eliminated OUD treatment waiting lists in all
counties, making treatment available on demand,b referring to the elimination of wait lists at hubs, but not
necessarily at spokes.c In addition, Vermont’s higher prescribing rate may also reflect higher treatment dosages.d As
indicated in the American Society of Addiction Medicine’s national guidelines, a typical dose of 16 mg of
buprenorphine is commonly taken as two 8 mg tablets (ASAM 2015), thus higher prescribing rates could suggest
higher dosing. Growing evidence suggests that higher doses of buprenorphine (e.g., 16–32 mg) are more efficacious
than lower doses (Mattick et al. 2014; Thomas et al. 2014).
a
”Hub-and-Spoke,” State of Vermont Blueprint for Health, accessed August 5, 2019,
https://blueprintforhealth.vermont.gov/about-blueprint/hub-and-spoke.
b
“Governor Phil Scott, State and Local Officials Announce Vermont Can Provide Opiate Addiction Treatment and Cure upon
Request,” State of Vermont Office of Governor Phil Scott, September 14, 2017, https://governor.vermont.gov/press-
release/governor-phil-scott-state-and-local-officials-announce-vermont-can-provide-opiate.
c
Lola Duffort, “Vermont’s Governor Inflated His Record on Reducing Opioid Treatment Waiting Lists,” Politicfact, November 8,
2018, https://www.politifact.com/vermont/statements/2018/nov/08/phil-scott/vermonts-governor-inflating-his-record-when-
it-com/; “Opioid Use Disorder Treatment Census and Wait List,” Vermont Department of Health, accessed August 5, 2019,
https://www.healthvermont.gov/sites/default/files/documents/pdf/ADAP_OpioidUseDisorderTreatmentCensusandWaitList.pdf
d
When we assessed monthly prescription patterns, we found that most people receiving buprenorphine prescriptions in Vermont
received multiple prescriptions per month, and a higher proportion in Vermont received multiple prescriptions per month than in
the other four states we assessed, which is consistent with Vermont’s dispensing higher treatment dosages.

Large differences in states’ prescribing rates are evident when examining these rates by decile. To
quantify these differences, we examined the ratio of the 90th- to 10th-percentile states. The 90th-
percentile state has the fifth-highest rate of buprenorphine maintenance treatment prescriptions per
enrollee, and the 10th-percentile state has the fifth-lowest prescribing rate. Using this metric, we found
that the highest buprenorphine maintenance treatment prescribing rate is 31 times greater than the
lowest prescribing rate in 2018 (table 4). This statistic is based on the ratio of two states’ prescribing
rates: Ohio, the 90th-percentile state with the fifth-highest prescribing rate (438) and Kansas, the 10th-
percentile state with the fifth-lowest prescribing rate of 14 (table 5). We also find that relative

12 STATE VARIATION IN MEDICAI D PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18

differences in buprenorphine maintenance treatment prescriptions per enrollee decreased between
2011 and 2018.

TABLE 4
Variation in Buprenorphine Maintenance Treatment Prescriptions per 1,000 Medicaid Enrollees Ages
12 and Older, All States and the District of Columbia, 2011 and 2018

10th- 90th-
Buprenorphine maintenance treatment Median percentile percentile Relative
prescriptions per 1,000 Medicaid enrollees state state state variation ratioa
2011 20 3 126 43
2018 82 14 438 31
Overall change, 2011–18b 317% 381% 247%
Source: Urban Institute analysis of Medicaid State Drug Utilization Data from the Centers for Medicare & Medicaid Services.
Notes: The 10th-percentile state has the fifth-lowest prescribing rate, and the 90th-percentile state has the fifth-highest
prescribing rate. The Medicaid State Drug Utilization Data exclude prescriptions written by prescribers at some safety net
providers participating in the 340B medication rebate program, such as federally funded clinics. See the methodological
appendices for more details (Clemans-Cope, Epstein, Winiski, et al. 2019; Lynch, Winiski, and Clemans-Cope 2019).
a
Relative state variation was calculated as the ratio of the 90th-percentile state to the 10th-percentile state.
b
Change is calculated as the percentage change from 2011 to the 2018 value.

Discussion
This brief presents new Medicaid prescription data to give policymakers and the public insight on how
states are leveraging Medicaid to address the opioid crisis. Our analysis of increases in buprenorphine
maintenance treatment prescriptions for OUD suggests a large-scale increase in access to treatment,
especially in expansion states, and particularly in a few states such as Vermont and West Virginia.

We found enormous differences across states in the number of buprenorphine maintenance

treatment prescriptions per Medicaid enrollee ages 12 and older. Relative differences in states’
prescribing rates may reflect differences in underlying rates of OUD, dosing, and prescribing patterns,
but it is unlikely that such factors account for the 31-fold state variation in prescriptions per enrollee we
find in 2018. That variation is far larger than the two- to threefold relative state differences in opioid
prescribing per capita (Schieber et al. 2019). Those differences in opioid prescribing rates attracted a
great deal of policy and public attention, but differences in buprenorphine maintenance treatment
prescribing across states have not yet been recognized.

Vermont’s rate of 1,210 buprenorphine maintenance treatment prescriptions per 1,000 Medicaid
enrollees suggests higher treatment rates than in other states. With an estimated treatment rate of 73
percent in 2014, before much of Vermont’s hub-and-spoke treatment system expansion (Brooklyn and
Sigmon 2017), current treatment rates could exceed 80 percent. In late 2017, the governor of Vermont
announced that treatment was available on demand after the elimination of OUD treatment waiting
lists in all counties,8 referring to the elimination of hub wait lists, and health department data show no
wait lists at hubs by 2018 and into 2019.9

STATE VARIATION IN M EDICAID PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18 13

 Despite Vermont’s high prescribing rate, Vermont officials do not report negative consequences
related to “diversion” of buprenorphine to people without a prescription. Buprenorphine is said to be
“diverted” from a person with a prescription to someone without a buprenorphine prescription, often
because that person had trouble accessing treatment. In March 2019, Vermont’s commissioner of the
Department of Public Safety said, “The idea that you have diverted buprenorphine as a self-treatment
option because of barriers to treatment, I just don’t think that holds water any longer in Vermont,
because we’ve done such a good job of making this kind of treatment available virtually on demand.”10
To the extent that diversion occurs, most buprenorphine use without a prescription is used for
“controlling withdrawal and cravings for other opioids and not to get high.”11 Burlington, Vermont,
Police Chief Brandon Del Pozo observed that “taking nonprescribed buprenorphine predicts an
increased eventual desire to enter into formal treatment.”12 Thus, in 2019, Vermont considered
legislation to decriminalize possession of buprenorphine without a prescription.13

Vermont’s high prescribing rate also suggests patients may be given a higher dose (e.g., 16 mg
buprenorphine taken as two 8 mg tablets). Growing evidence suggests higher doses of buprenorphine
(e.g., 16–32 mg) are more efficacious than lower doses (Mattick et al. 2014; Thomas et al. 2014), so
Vermont’s higher prescribing rates could reflect clinically effective dosing.

Though more research is needed to determine whether treatment decreased opioid-related deaths,
Vermont had the lowest opioid-related death rate in 2017 across New England states. In contrast, rates
of opioid-related overdose deaths remain high in some Medicaid expansion states, such as West
Virginia. Our estimates show West Virginia’s prescribing rate, the second highest in the nation, is far
below Vermont’s, suggesting current treatment levels in West Virginia may not meet underlying need.

Florida, with an opioid-related overdose death rate above the national average, has the fourth-
lowest per enrollee buprenorphine maintenance treatment prescribing rate in 2018. Several states,
including Florida, have not expanded Medicaid and have opioid-related death rates higher than the
national average in 2018 (Missouri, North Carolina, South Carolina, Tennessee, Utah, and Wisconsin).14
Of these, Missouri, South Carolina, Tennessee, Utah, and Wisconsin were in the lower half of rates of
buprenorphine maintenance treatment prescriptions per Medicaid enrollee in 2018. Medicaid
expansion in these states would likely increase the share of people with access to both overdose
reversal and treatment drugs, potentially increasing provider capacity and access to treatment and
providers.

Yet, even many states that expanded Medicaid and have relatively high buprenorphine
maintenance treatment prescribing rates could gain from understanding the factors underlying the
differences between their rate and Vermont’s, the highest rate of Medicaid-covered buprenorphine
maintenance treatment prescriptions in the United States.

This analysis is only a first step and raises important questions. States will want to assess the
effectiveness of treatments offered by providers in their state, including whether higher buprenorphine
dosing and on-demand treatment upon demand increase retention and recovery. Further research
could use claims data to assess characteristics of pharmacotherapy for OUD, including prescription

14 STATE VARIATION IN MEDICAI D PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18

dosage, duration, and formulation. For example, to what extent do state Medicaid programs differ in the
dosage, duration, and formulation of buprenorphine treatment prescriptions for those in treatment? Do
differences reflect patient characteristics, including the patient’s phase of treatment, or something else?
To what extent do observed treatment patterns reflect clinical guidelines? It will also be critical to
examine the extent to which differences in pharmacotherapy across and within states are associated
with differences in mortality and claims-based outcomes, such as medication continuation, emergency
department visits, or hospitalizations related to drug use. It is important that claims analyses be
supplemented by discussions with buprenorphine treatment prescribers to understand what shapes
their treatment decisions and, particularly, what roles dosage limits, utilization review, and payment and
billing practices play, and how those roles vary across states and payers.

Finally, in many places, receipt of evidence-based treatment and recovery services for OUD, such as
buprenorphine maintenance treatment, may be hampered by factors such as stigma, low rates of
screening and referral, lack of provider capacity, and prescribing restrictions. However, several state
models for expanding access to OUD treatment in Medicaid are being pursued.15 Closing coverage gaps
in states that have not expanded Medicaid and addressing demand and supply factors will likely be
needed to reduce harm caused by untreated OUD.

Limitations
Our study has several limitations. First, our analysis excludes other types of OUD pharmacotherapy:
methadone and naltrexone. Second, the Medicaid SDUD exclude claims representing drugs purchased
under the 340B Drug Pricing Program by certain safety net providers, including federally funded clinics,
and thus undercount Medicaid-covered prescriptions overall. To the extent states differ in their shares
of buprenorphine treatment prescriptions provided through entities participating in the 340B
medication rebate program, state rankings may differ from the patterns shown here. Third, we cannot
distinguish prescriptions written for Medicaid enrollees newly gaining coverage under the expansion
from those written for existing enrollees. Fourth, the per enrollee estimates might not reflect treatment
relative to need, because need for OUD treatment varies across states, as does access to methadone for
OUD, which can substitute for buprenorphine OUD treatment. In addition, per enrollee estimates are
derived from aggregate data, not individual-level data, and thus are a rough measure of prescriptions
per person. Fifth, this analysis considers medication treatments only, though many people with OUD
need more intensive treatment and services and recovery support, as well as treatment for
comorbidities common among people with OUD, such as mental health problems, hepatitis C, and
HIV/AIDS. See the study methodology appendix for a more detailed description of the limitations
(Clemans-Cope, Epstein, Winiski, et al. 2019).

STATE VARIATION IN M EDICAID PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18 15

TABLE 5
Buprenorphine Maintenance Treatment, Overall and per 1,000 Medicaid Enrollees Ages 12 and
Older, by State, 2011, 2014, and 2018
Ordered by largest to smallest 2018 per enrollee estimates

BMT Prescriptions per 1,000 Medicaid

Total BMT Prescriptions Enrollees Ages 12 and Older
State 2011 2014 2018 2011 2014 2018
VT 66,661 106,636 157,872 590.9 784.1 1,209.8
WV 36,799 92,851 328,543 143.9 210.1 826.6
KY 48,867 24,789 659,629 92.9 28.2 662.2
MT 1,236 1,565 95,598 19.8 21.5 587.5
OH 82,412 249,797 943,897 54.9 116.4 437.7
ME 59,278 60,966 88,205 206.8 281.0 403.5
MA 150,483 259,136 457,909 175.4 172.8 384.3
DE 13,899 19,240 58,869 93.4 127.8 291.2
MD 76,525 118,693 261,702 126.2 153.6 283.5
IN 20,730 32,384 229,527 31.8 53.5 256.2
NH 7,906 15,302 32,008 81.5 147.8 246.1
OR 5,395 54,891 183,096 15.8 58.2 236.1
PA 141,148 177,991 557,288 89.0 108.7 224.3
VA 33,528 40,977 134,333 60.8 79.9 217.5
RI 12,101 27,876 45,934 93.9 155.9 207.5
CT 55,525 80,269 150,207 104.7 131.2 186.9
AK 3,740 6,265 25,199 45.1 81.8 182.8
WA 8,494 26,844 224,991 11.4 18.6 145.5
NC 33,787 74,873 172,372 33.4 70.8 139.2
NM 8,195 32,274 72,765 28.5 78.0 139.2
WI 28,101 42,536 85,927 34.4 48.8 108.5
LA 13,528 19,293 83,207 21.8 29.1 99.3
CO 3,003 13,636 84,638 7.2 21.9 97.6
NJ 24,812 57,016 116,755 40.5 47.9 88.2
MI 22,227 32,489 145,096 19.0 17.3 84.9
NY 169,077 243,213 448,596 45.4 49.0 82.5
ND 699 732 5,884 13.8 12.2 81.6
AL 5,720 12,011 31,301 12.4 23.1 65.1
AZ 18,902 15,289 78,361 22.9 12.3 64.0
MO 9,902 13,034 40,468 16.3 20.3 63.0
UT 5,595 5,062 10,979 32.3 24.7 62.3
DC 3,680 7,079 12,312 21.7 36.3 58.4
SC 7,542 12,578 32,233 14.5 19.9 51.9
ID 1,838 3,017 8,289 13.9 18.7 49.3
MN 11,349 21,468 42,954 16.5 26.3 48.8
IL 22,251 11,368 97,733 12.3 5.6 47.3
TN 50,272 75,366 32,337 61.8 89.4 35.1
OK 4,615 7,481 15,101 9.4 16.7 33.7
MS 3,831 6,635 12,061 10.4 16.5 33.5
IA 1,569 2,598 16,398 4.5 5.8 31.4
HI 2,976 5,219 8,229 15.1 23.0 31.4
NV 840 2,825 13,550 4.6 8.1 29.0
WY 591 677 936 14.4 15.6 24.4

16 STATE VARIATION IN MEDICAI D PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18

 BMT Prescriptions per 1,000 Medicaid
Total BMT Prescriptions Enrollees Ages 12 and Older
State 2011 2014 2018 2011 2014 2018
CA 10,025 27,395 169,794 2.1 3.3 18.5
NE 75 217 2,209 0.5 1.4 15.2
GA 3,330 10,522 15,405 3.7 10.3 14.8
KS 555 2,966 3,157 2.6 13.0 14.1
FL 7,473 16,646 36,132 3.8 7.2 12.3
SD 196 329 725 2.9 4.7 10.8
TX 10,155 13,143 18,996 4.8 6.2 8.4
AR 654 2,039 3,385 2.1 2.6 5.4
US 1,312,051 2,187,434 6,428,482 36.2 45.2 121.8
Sources: Urban Institute analysis of Medicaid State Drug Utilization Data from the Centers for Medicare & Medicaid Services and
Multiple Cause of Death, 1999–2017, data from the Centers for Disease Control and Prevention WONDER online database,
released December 2018 and available at https://wonder.cdc.gov/mcd-icd10.html. Data from the Multiple Cause of Death, 1999–
2017, files were compiled from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative
Program.
Notes: BMT = buprenorphine maintenance treatment. All buprenorphine maintenance treatment prescriptions and those per
1,000 Medicaid enrollees ages 12 and older were derived from all buprenorphine maintenance treatment prescriptions in
Medicaid State Drug Utilization Data for each state and the District of Columbia in that year. The Medicaid State Drug Utilization
Data exclude prescriptions written by prescribers at some safety net providers participating in the 340B medication rebate
program, such as federally funded clinics. See the methodological appendices for more details (Clemans-Cope, Epstein, Winiski, et
al. 2019; Lynch, Winiski, and Clemans-Cope 2019).

Notes
1 Outcomes of buprenorphine treatment for OUD include decreased mortality (Degenhardt et al. 2009; Schwartz et
al. 2013; Sordo et al. 2017); reduced morbidity (Romelsjö et al. 2010), including reduced HIV and hepatitis C
infection (Lawrinson et al. 2008; Tsui et al. 2014); increased retention in OUD treatment (Mattick et al. 2014);
decreased relapse events, including hospitalizations and emergency department visits related to overdose (Clark
et al. 2011); and reduced involvement with the justice system (Dunlop et al. 2017). Methadone treatment for
OUD is also highly effective, as is naltrexone, particularly with highly motivated patients (Schuckit 2016).
Naloxone has been shown to be highly effective at and safe for reducing opioid overdose mortality (Wermeling
2015).
2 See note 1.

3 For most states we examined, the most recent quarter of MSIS data was the fourth quarter of 2015.
4 “Medicaid Statistical Information Statistics (MSIS),” Centers for Medicare & Medicaid Services, updated April 3,
2017, https://www.cms.gov/Research-Statistics-Data-and-Systems/Computer-Data-and-Systems/MSIS/;
“Medicaid Analytic eXtract (MAX) General Information,” Centers for Medicare & Medicaid Services, updated
March 14, 2019, https://www.cms.gov/Research-Statistics-Data-and-Systems/Computer-Data-and-
Systems/MedicaidDataSourcesGenInfo/MAXGeneralInformation.html.
5
See also “Status of State Action on the Medicaid Expansion Decision,” Henry J. Kaiser Family Foundation, updated
August 1, 2019, https://www.kff.org/health-reform/state-indicator/state-activity-around-expanding-medicaid-
under-the-affordable-care-act/; Lynn A. Blewett, “Medicaid ‘Early Opt-In’ States,” State Health Access Data
Assistance Center, October 9, 2015, https://www.shadac.org/news/medicaid-%E2%80%9Cearly-opt-
in%E2%80%9D-states.
6 When we assessed monthly prescription patterns, we found most people receiving buprenorphine prescriptions in
Vermont received multiple prescriptions per month, and a higher proportion in Vermont received multiple

STATE VARIATION IN M EDICAID PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18 17

 prescriptions per month than in the other four states we assessed, which, as indicated in box 1, is consistent with
Vermont’s dispensing higher treatment dosages.
7 Authors’ calculations based on Multiple Cause of Death, 1999–2017, data from the Centers for Disease Control

and Prevention WONDER online database, released December 2018 and available at
https://wonder.cdc.gov/mcd-icd10.html. Data from the Multiple Cause of Death, 1999–2017, files were
compiled from data provided by the 57 vital statistics jurisdictions through the Vital Statistics Cooperative
Program.
8 “Governor Phil Scott, State and Local Officials Announce Vermont Can Provide Opiate Addiction Treatment and

Care upon Request,” State of Vermont Office of Governor Phil Scott, September 14, 2017,
https://governor.vermont.gov/press-release/governor-phil-scott-state-and-local-officials-announce-vermont-
can-provide-opiate.
9 Lola Duffort, “Vermont’s Governor Inflated His Record on Reducing Opioid Treatment Waiting Lists,” Politifact,

November 8, 2018, https://www.politifact.com/vermont/statements/2018/nov/08/phil-scott/vermonts-

governor-inflating-his-record-when-it-com/; “Opioid Use Disorder Treatment Census and Wait List,” Vermont
Department of Health, accessed August 5, 2019,
https://www.healthvermont.gov/sites/default/files/documents/pdf/ADAP_OpioidUseDisorderTreatmentCensu
sandWaitList.pdf.
10 Peter Hirschfeld, “Health Experts Push to Decriminalize Addiction Treatment Drug,”
VPR, March 22, 2019,
https://www.vpr.org/post/health-experts-push-decriminalize-addiction-treatment-drug#stream/0.
11 “What Is the Treatment Need Versus the Diversion Risk for Opioid Use Disorder Treatment?” National Institute

on Drug Abuse, accessed August 8, 2019, https://www.drugabuse.gov/publications/medications-to-treat-

opioid-addiction/what-treatment-need-versus-diversion-risk-opioid-use-disorder-treatment.
12 Peter Hirschfeld, “Health Experts Push to Decriminalize Addiction Treatment Drug,” VPR.

13 H.162, 2019 Gen. Assemb., Reg. Sess. (Vt. 2019).

14 See note 7.
15 Tyler Sadwith, John O’Brien, Vikki Wachino, and Sarah Twardock, “Leveraging Medicaid to Combat the Opioid
Epidemic: How Leader States and Health Plans Deliver Evidence-Based Treatment,” Health Affairs Blog, June
24, 2019, https://www.healthaffairs.org/do/10.1377/hblog20190619.49397/full/.

References
ASAM (American Society of Addiction Medicine). 2015. The ASAM National Practice Guideline for the Use of
Medications in the Treatment of Addiction Involving Opioid Use. Chevy Chase, MD: American Society of Addiction
Medicine.
Brooklyn, John, and Stacey Sigmon. 2017. “Vermont Hub-and-Spoke Model of Care for Opioid Use Disorder:
Development, Implementation, and Impact.” Journal of Addiction Medicine 11 (4): 286–92.
Clark, Robin E., Mihail Samnaliev, Jeffrey D. Baxter, and Gary Y. Leung. 2011. “The Evidence Doesn’t Justify Steps
by State Medicaid Programs to Restrict Opioid Addiction Treatment with Buprenorphine.” Health Affairs 30 (8):
1425–33.
Clemans-Cope, Lisa, Marni Epstein, Victoria Lynch, and Emma Winiski. 2019. “Rapid Growth in Medicaid Spending
and Prescriptions to Treat Opioid Use Disorder and Opioid Overdose from 2010 to 2017.” Washington, DC:
Urban Institute.
Clemans-Cope, Lisa, Marni Epstein, Emma Winiski, and Victoria Lynch. 2019. “Tracking Medicaid-Covered
Prescriptions to Treat Opioid Use Disorder: Methodology Appendix.” Washington, DC: Urban Institute.
Clemans-Cope, Lisa, Victoria Lynch, Marni Epstein, and Genevieve M. Kenney. 2017. “Medicaid Coverage of
Effective Treatment for Opioid Use Disorder: Trends in State Buprenorphine Prescriptions and Spending Since
2011.” Washington, DC: Urban Institute.

18 STATE VARIATION IN MEDICAI D PRESCRIPTIONS FOR OPIOID USE DISORDER, 2011–18

———. 2019. “Opioid and Substance Use Disorder and Receipt of Treatment among Parents Living with Children in
the United States, 2015–2017.” Annals of Family Medicine 17 (3): 207–11.
Clemans-Cope, Lisa, Jane B. Wishner, Eva H. Allen, Nicole Lallemand, Marni Epstein, and Brenda C. Spillman. 2017.
“Experiences of Three States Implementing the Medicaid Health Home Model to Address Opioid Use Disorder—
Case Studies in Maryland, Rhode Island, and Vermont.” Journal of Substance Abuse Treatment 83: 27–35.
Degenhardt, Louisa, Deborah Randall, Wayne Hall, Matthew Law, Tony Butler, and Lucy Burns. 2009. “Mortality
among Clients of a State-Wide Opioid Pharmacotherapy Program over 20 Years: Risk Factors and Lives Saved.”
Drug and Alcohol Dependence 105 (1): 9–15.
Dunlop, Adrian J., Amanda L. Brown, Christopher Oldmeadow, Anthony Harris, Anthony Gill, Craig Sadler, et al.
2017. “Effectiveness and Cost-Effectiveness of Unsupervised Buprenorphine-Naloxone for the Treatment of
Heroin Dependence in a Randomized Waitlist Controlled Trial.” Drug and Alcohol Dependence 174: 181–91.
Grogan, Colleen M., Christina Andrews, Amanda Abraham, Keith Humphreys, Harold A. Pollack, Bikki Tran Smith,
et al. 2016. “Survey Highlights Differences in Medicaid Coverage for Substance Use Treatment and Opioid Use
Disorder Medications.” Health Affairs 35 (12): 2289–96.
Lawrinson, Peter, Robert Ali, Aumphornpun Buavirat, Sithisat Chiamwongpaet, Sergey Dvoryak, Bouslaw Habrat,
et al. 2008. “Key Findings from the WHO Collaborative Study on Substitution Therapy for Opioid Dependence
and HIV/AIDS.” Addiction 103 (9): 1484–92.
Lynch, Victoria, Emma Winiski, and Lisa Clemans-Cope. 2019. “Medicaid Enrollment for Individuals Who May
Receive a Prescription to Treat Opioid Use Disorder.” Washington, DC: Urban Institute.
Maclean, Johanna Catherine, and Brendan Saloner. 2017. “The Effect of Public Insurance Expansions on Substance
Use Disorder Treatment: Evidence from the Affordable Care Act.” NBER Working Paper 23342. Cambridge, MA:
National Bureau of Economic Research.
MACPAC (Medicaid and CHIP Payment and Access Commission). 2017. “Medicaid and the Opioid Epidemic.” In
Report to Congress on Medicaid and CHIP. Washington, DC: Medicaid and CHIP Payment and Access Commission.
———. 2018. “The 340B Drug Pricing Program and Medicaid Drug Rebate Program: How They Interact.”
Washington, DC: Medicaid and CHIP Payment and Access Commission.
Mattick, Richard P., Courtney Breen, Jo Kimber, and Marina Davoli. 2014. “Buprenorphine Maintenance Versus
Placebo or Methadone Maintenance for Opioid Dependence.” Cochrane Database of Systematic Reviews (2): 1–85.
Miller, Niki. 2018. Comprehensive Update on State Medicaid Coverage of Medication-Assisted Treatments and Substance
Use Disorder Services. Germantown, MD: Advocates for Human Potential Inc.
Murrin, Suzanne. 2016. State Efforts to Exclude 340B Drugs from Medicaid Managed Care Rebates. OEI-05-14-00430.
Washington, DC: Department of Health and Human Services, Office of the Inspector General.
Romelsjö, Anders P., Barbo Engdahl, Marlene Stenbacka, Anna B. Fugelstad, Ingrid Davstad, Anders G. Leifman, et
al. 2010. “Were the Changes to Sweden’s Maintenance Treatment Policy 2000-06 Related to Changes in Opiate-
Related Mortality and Morbidity?” Addiction 105 (9): 1625–32.
Saloner, Brendan, and Shankar Karthikeyan. 2015. “Changes in Substance Abuse Treatment Use among Individuals
with Opioid Use Disorders in the United States, 2004–2013.” JAMA 314 (14): 1515–17.
SAMHSA (Substance Abuse and Mental Health Services Administration). 2018. Medicaid Coverage of Medication-
Assisted Treatment for Alcohol and Opioid Use Disorders and of Medication for the Reversal of Opioid Overdose.
Rockville, MD: Substance Abuse and Mental Health Services Administration.
Schieber, Lyna Z., Gery P. Guy, Puja Seth, Randall Young, Christine L. Mattson, Christina A. Mikosz, et al. 2019.
“Trends and Patterns of Geographic Variation in Opioid Prescribing Practices by State, United States, 2006–
2017.” JAMA Network Open 2 (3): e190665.
Schuckit, Marc A. 2016. “Treatment of Opioid-Use Disorders.” New England Journal of Medicine 375 (4): 357–68.
Schwartz, Robert P., Jan Gryczynski, Kevin E. O’Grady, Joshua M. Sharfstein, Gregory Warren, Yngvild Olsen, et al.
2013. “Opioid Agonist Treatments and Heroin Overdose Deaths in Baltimore, Maryland, 1995–2009.” American
Journal of Public Health 103 (5): 917–22.

BRIEF TITLE HERE: USE A SHORTENED TITLE IF N ECESSARY 19

Sommers, Benjamin D., Emily Arntson, Genevieve M. Kenney, and Arnold M. Epstein. 2013. “Lessons from Early
Medicaid Expansions under Health Reform: Interviews with Medicaid Officials.” Medicare & Medicaid Research
Review 3 (4): E1–E23.
Sordo, Luis, Gregorio Barrio, Maria J. Bravo, B. Iciar Indave, Louisa Degenhardt, Lucas Wiessing, et al. 2017.
“Mortality Risk during and after Opioid Substitution Treatment: Systematic Review and Meta-Analysis of Cohort
Studies.” BMJ 357: j1550.
Thomas, Cindy Parks, Catherine Anne Fullerton, Meelee Kim, Leslie Montejano, Russell Lyman, Richard H.
Dougherty, et al. 2014. “Medication-Assisted Treatment with Buprenorphine: Assessing the Evidence.”
Psychiatric Services 65 (2): 158–70.
Tsui, Judith I., Jennifer L. Evans, Paula J. Lum, Judith A. Hahn, and Kimberly Page. 2014. “Association of Opioid
Agonist Therapy with Lower Incidence of Hepatitis C Virus Infection in Young Adult Injection Drug Users.” JAMA
Internal Medicine 174 (12): 1974–81.
Wen, Heifei, Jason Hockenberry, Tyrone Borders, and Benjamin Druss. 2017. “Impact of Medicaid Expansion on
Medicaid-Covered Utilization of Buprenorphine for Opioid Use Disorder Treatment.” Medical Care 55 (4): 336–
41.
Wermeling, Daniel P. 2015. “Review of Naloxone Safety for Opioid Overdose: Practical Considerations for New
Technology and Expanded Public Access.” Therapeutic Advances in Drug Safety 6 (1): 20–31.

About the Authors

Lisa Clemans-Cope is a principal research associate and health economist in the Health Policy Center at
the Urban Institute. Her areas of expertise include substance use disorder treatment, health spending,
access to and use of health care, private insurance, Medicaid and CHIP programs, dual eligibles, health
reform legislation and regulation, and health-related survey and administrative data. She has led
qualitative and quantitative research projects examining the impacts of policies aimed at improving
diagnosis and treatment of people with substance use disorders. Clemans-Cope holds a BA in
economics from Princeton University and a doctorate in health economics from the Johns Hopkins
Bloomberg School of Public Health.

Victoria Lynch is a senior research associate in the Health Policy Center. She is a survey methodologist
with in-depth understanding of public policy on Medicaid, the Children’s Health Insurance Program, and
other health insurance, as well as extensive experience with Medicaid administrative claims data. She
received her MS in survey methodology from the Joint Program in Survey Methodology, run by the
University of Maryland, University of Michigan, and Westat.

Emma Winiski is a research assistant in the Health Policy Center. Her current works focuses on
treatment of substance use disorder under Medicaid and analysis of survey data. She received a BS from
Furman University.

Marni Epstein is a research assistant in the Health Policy Center. Her current works focuses on
treatment of substance use disorder under Medicaid and quantitative analysis of Medicaid
administrative claims data. She received a BA from Johns Hopkins University.

20 BRIEF TITLE HERE: US E A SHORTENED TITLE IF NECESSARY

Acknowledgments
This brief was funded by the Laura and John Arnold Foundation. We are grateful to them and to all our
funders, who make it possible for Urban to advance its mission.

The views expressed are those of the authors and should not be attributed to the Urban Institute,
its trustees, or its funders. Funders do not determine research findings or the insights and
recommendations of Urban experts. Further information on the Urban Institute’s funding principles is
available at urban.org/fundingprinciples.

The authors thank Kimá Joy Taylor, Genevieve M. Kenney, and Vikki Wachino for helpful comments
and suggestions.

ABOUT THE URBAN INST ITUTE

The nonprofit Urban Institute is a leading research organization dedicated to
developing evidence-based insights that improve people’s lives and strengthen
communities. For 50 years, Urban has been the trusted source for rigorous analysis
of complex social and economic issues; strategic advice to policymakers,
philanthropists, and practitioners; and new, promising ideas that expand
opportunities for all. Our work inspires effective decisions that advance fairness and
500 L’Enfant Plaza SW
enhance the well-being of people and places.
Washington, DC 20024
Copyright © August 2019. Urban Institute. Permission is granted for reproduction
www.urban.org
of this file, with attribution to the Urban Institute.

BRIEF TITLE HERE: USE A SHORTENED TITLE IF N ECESSARY 21